1. Field of the Invention
The present invention generally relates to a server rack system, in particular, to a server rack system for managing power supply.
2. Description of Related Art
A server has high computing capacity, and is thus used as a core computer that serves the other computers in a network system. The server, functioning as the core computer that serves the other computers in the network system, is capable of providing functions such as a magnetic disk and a printing service required by a network user, and is also capable of enabling clients to share with each other various resources within the network environment.
The servers are generally classified, according to their appearances, into three types, including a pedestal server, a rack server and a blade server. The rack server is a pedestal server that is optimized in structure, and designed to reduce an occupied space of the server. The width of the rack server is usually 19 inches, and the height thereof is measured with a unit U (1 U=1.75 inches). Taking a server rack for example, the server rack is a server with its appearance designed according to a uniform standard, and is used in combination with a cabinet. In other words, the rack server is a tower server that is optimized in structure, and mainly designed to reduce an occupied space of the server as much as possible. Many professional network apparatuses adopt the rack type structure, and are mostly in a flat shape like a drawer, such as a switch, a router, and a hardware firewall. The width of the server rack is 19 inches, and the height thereof is measured with a unit U (1 U=1.75 inches=44.45 millimeters). Usually, there are servers in standards of 1 U, 2 U, 3 U, 4 U, 5 U, and 7 U.
The basic architecture of a server is generally the same as a personal computer, which includes components such as a central processing unit (CPU), a memory, and an input/output device. The components are connected through buses within the server, and are connected through a north bridge chip to the CPU and the memory and connected through a south bridge chip to the input/output device. However, sufficient electric power supply is required in order to enable all the components to work normally. When an error occurs in the configuration of a power supply unit (PSU) that is in charge of electric power supply in the server, the quality of the server is affected, and circumstances such as difficulty in startup, a set value error, or system unsteadiness may occur.
Generally speaking, for server units located in a rack (such as the switch, the router, and the hardware firewall), cables are generally used to connect each server unit and a power supply module, so that the power supply module provides desired electric power for each server unit. However, in this case, an extra space is required in the rack to place the cables and allow the cables to move accordingly with the cabinet when the server unit is drawn out. Thereby, it is easy to cause poor space allocation and utilization rate in the rack. In addition, with the increasing of the number of the server units, the number of the cables increases accordingly, and the disarrangement of the cables may hinder the operation of a user.
Furthermore, the servers are usually placed in special machine rooms, and power control and management is required to be performed in a remote manner, so that the PSU must have a remote management function. In other words, the PSU needs to use a network power. However, the use of the network power leads to the problem of unsteadiness of the network power and causes dramatic increase of the hardware cost.